SSRI antidepressants: Information for health professionals concerning use of SSRI antidepressants in pregnant women

Related information

7 September 2005

New information has recently become available that suggests an association between use of SSRI antidepressant medicines in early pregnancy and congenital heart abnormalities. This association appears to be strongest for paroxetine.

Nature of the problem

In a population-based cohort study, Danish researchers have reported an association between use of SSRI antidepressants in the first trimester of pregnancy and cardiac abnormalities. The odds ratio for this was 1.6 (95%CI: 1.1-1.9), suggesting a 60% higher chance of developing a cardiac abnormality.

The authors have noted that the analysis does not prove that SSRI use caused the abnormalities, and there could be other unknown factors affecting the result. No subgroup analysis on the prevalence of cardiac abnormalities in the infants of women taking specific SSRIs is available from this study at present.

Preliminary analysis of data from another study, conducted in the USA, suggests that the use of paroxetine may be associated with an increased risk of congenital malformations and cardiovascular malformations. In this study, infants born to three groups of women were compared: women given bupropion during the first trimester of pregnancy, women given bupropion before or after the first trimester of pregnancy and women given other antidepressants during the first trimester of pregnancy.

In this study, compared with infants of mothers given other antidepressants, infants of mothers given paroxetine in the first trimester of pregnancy were 2.2 times more likely to have a congenital malformation and 2.08 times more likely to have a cardiovascular malformation. The overall incidence of congenital malformations in infants whose mothers were given paroxetine was approximately 4%. The incidence of cardiovascular malformations was approximately 2%. The analysis of this study is preliminary and further analysis is required. The majority of the abnormalities in infants exposed to paroxetine were ventricular septal defects. Infants with chromosomal defects were not considered as cases of malformation in this analysis.

Paroxetine is registered in Australia under the trade names Aropax, Oxetine, Paxtine, Chem mart, GenRx and Terry White Chemists paroxetine, Paroxetine-RL, Paroxetine-BC, Paroxetine Hexal, Espar, Loxamine, Paroxat CR and Ausrox.

The preliminary study results compare to a prevalence in the general population of approximately 3% for birth defects overall and 0.5-1.0% for heart defects. From the table it is clear that too few infants were exposed to some antidepressants to make recommendations based on these data. However over 500 pregnant women in the first trimester have been exposed to each of fluoxetine, paroxetine and sertraline. The evidence for these medicines is more robust. Of these sertraline was associated with a lower incidence of cardiac abnormalities compared with exposure to other antidepressants. It must be emphasised that there is no comparator group in this study in which infants were not exposed to any antidepressant. This does not prove sertraline does not cause congenital malformations but there is no strong potential safety signal as with paroxetine.


Women who are taking paroxetine and are pregnant or are planning to become pregnant should have their treatment reviewed.

Women should be advised that while use of paroxetine is not proven to cause abnormalities, it would be prudent not to take it while intending to become pregnant or in the first thirteen weeks of pregnancy.

Patients should not suddenly stop taking paroxetine as they may have withdrawal effects that can be severe or life-threatening. Dosage must be tapered gradually as per the Product Information instructions.

The most common cardiac abnormality seen in the study was ventricular septal defect (VSD). This can also occur without any medicine exposure and is rarely life-threatening. It can resolve spontaneously in some cases.

In the US study some medicines were taken by large numbers of women without an apparent increase in risk of malformation. These medicines included sertraline.

The following table shows the odds ratios for cardiovascular malformation according to mutually exclusive categories of specific antidepressants dispensed during the first trimester.

  Cases with known use of teratogens excluded   Ever use cases including use of known teratogens
Antidepressant Cases* No. of infants Adjusted‡
Odds Ratio
95% confidence interval   Cases** No. of infants Adjusted
Odds Ratio
95% confidence interval
Amitriptyline 1 171 0.55 0.08-4.16   1 233 0.37 0.05-2.73
Amitriptyline/ Chlordiazepoxide 0 3 0     0 5 0  
Bupropion 2 282 0.64 0.15-2.66   8 463 1.58 0.72-3.46
Citalopram 5 217 2.22 0.83-5.95   6 2988 1.74 0.70-4.31
Clomipramine 0 3 0     0 5 0  
Desipramine 0 5 0     0 10 0  
Doxepin 0 19 0     0 22 0  
Fluoxetine 12 955 1.23 0.62-2.44   17 1178 1.45 0.78-2.70
Fluvoxamine 0 17 0     0 26 0  
Imipramine 0 29 0     0 42 0  
Mirtazapine 0 8 0     0 23 0  
Nefazodone 0 50 0     0 75 0  
Nortriptyline 0 69 0     0 87 0  
Paroxetine 11 589 2.08 1.03-4.23   14 704 2.26 1.17-4.33
Protriptyline 0 4 0     0 4 0  
Sertraline 0 563 0     1 705 0.09 0.01-0.67
Trazodone 1 58 1.66 0.22-12.43   1 154 0.51 0.07-3.91
Venlafaxine 2 150 1.22 0.29-5.15   3 215 1.18 0.36-3.86
>1 antidepressant 6 476 1.19 0.49-2.86          

*Excluded maternal dispensing of a teratogenic drug known or suspected to affect the cardiovascular system within 3 months before pregnancy

**Included cases with ever use of an antidepressant during the first trimester and maternal dispensing of a teratogenic drug known or suspected to affect the cardiovascular system within 3 months before pregnancy

†confidence intervals ranging from <1 to >1 do not suggest a statistically valid result

‡Adjusted for age, calendar year of delivery, diagnosis of pre-eclampsia or eclampsia and infant sex.

Top of page